A High-Precision Prediction Method for Local Hot Spot Temperature Based on Compensating Heat Source

Abstract

The present invention discloses a high-precision prediction method for local hot spot temperature based on a compensated heat source, which relates to the technical field of motor simulation modeling. First, based on the steady-state heat conduction equation, a local hot spot temperature including a nonlinear term is constructed. The least squares method is used to approximately express the temperature extreme value as a linear superposition of boundary conditions and heat sources. Further, a compensated heat source term is introduced to correct the nonlinear error, and a high-precision local hot spot prediction model is obtained. The present invention accurately corrects the fitting error generated during the nonlinear-to-linear conversion of the local hot spot temperature through a compensation mechanism, improves the prediction accuracy of the local hot spot temperature, maintains the structural simplicity of the thermal network model, greatly reduces the complexity of the heat conduction problem, and improves the calculation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of motor simulation modeling, and particularly to a high-precision prediction method for local hot spot temperature based on a compensated heat source. Background Art

[0002] Low-speed high-torque permanent magnet synchronous motors have become the core technology in fields such as heavy machinery, ship propulsion, and mining equipment due to their high torque output and high efficiency characteristics under low-speed operating conditions. In these applications, motors usually need to operate for a long time under extreme loads and complex operating conditions. Therefore, ensuring their stable operation within a safe temperature range is crucial. However, low-speed high-torque permanent magnet synchronous motors face severe heat dissipation challenges during operation. Due to the low thermal conductivity of air and the weak convection effect inside low-speed motors, heat is difficult to dissipate effectively, resulting in a sharp increase in the local temperature of the end windings and the formation of local hot spots. Research shows that for every 10°C increase in winding temperature, the service life of the insulation material will be shortened by approximately 50%. In severe cases, it may even lead to motor failures or safety accidents. Currently, research on motor cooling technology has received extensive attention, and the accurate estimation of motor temperature, especially the prediction of hot spot temperatures of windings and permanent magnets, is equally important or even more important. Therefore, establishing an accurate and reliable thermal model to accurately evaluate the hot spot temperature distribution of motors has become a key prerequisite for improving the torque density and operating reliability of permanent magnet motors.

[0003] Thermal modeling methods are mainly divided into two categories: numerical calculation methods and lumped parameter thermal network methods. Numerical calculation methods achieve temperature field analysis by solving partial differential equations, which require high computing resources and long computing time. In contrast, the thermal network model transforms the problem of solving partial differential equations of the temperature field into the problem of solving differential equations of a thermal circuit network, significantly reducing the complexity of temperature field analysis, with high computing efficiency and short computing time, making it the preferred method in motor thermal analysis. However, when there is a large temperature gradient in the system, the linearization assumption of thermal resistance in the lumped parameter thermal network model ignores the non-linear characteristics of thermal conductivity changing with temperature, introducing systematic errors and reducing the accuracy of temperature calculation. At the same time, the lumped parameter thermal network model based on the averaging assumption masks local temperature extremes, and its prediction result is the average temperature and cannot capture the maximum temperature, which may lead to undetected local overheating. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a high-precision prediction method for the local hot spot temperature based on a compensation heat source. Based on the principle of thermal steady-state balance, a linear mathematical mapping model between the local hot spot temperature, boundary conditions, and heat source distribution characteristics is constructed through theoretical derivation. Further, the compensation heat source is introduced into the traditional thermal network model in the form of an equivalent voltage source, and by adjusting the intensity of the compensation heat source, the prediction error caused by non-linear factors in the linearization process of the local hot spot temperature is effectively corrected. The present invention not only significantly improves the prediction accuracy of the local hot spot temperature, but also greatly reduces the complexity of the heat conduction problem and improves the calculation efficiency. The specific scheme is as follows:

[0005] The present invention discloses a high-precision prediction method for the local hot spot temperature based on a compensation heat source, including the following steps:

[0006] Build a finite element electromagnetic simulation model of the motor, and calculate the core loss, winding copper loss, and permanent magnet eddy current loss of the motor;

[0007] According to the thermal conductivity of the material, the motor is divided and equivalent into N cylindrical components to obtain equivalent cylindrical components; N>1; according to the heat transfer mode of the motor components, the connection thermal resistance between each component is determined;

[0008] Based on each equivalent cylindrical component, build a local hot spot temperature prediction model with a compensation heat source;

[0009] Improve the local hot spot prediction model of the motor, solve the steady-state heat balance equation, calculate the node temperature, and obtain the local hot spot temperature of the corresponding component.

[0010] Further, the motor is divided and equivalent into N cylindrical components according to the thermal conductivity of the material. Specifically, the motor is divided into a stator core, a permanent magnet, a rotor core, and a shaft according to the thermal conductivity of the material and is equivalent to cylindrical components.

[0011] Further, the construction of the local hot spot temperature prediction model with a compensation heat source specifically includes:

[0012] By performing differential operations on the general solution of the radial steady-state heat conduction equation, the local hot spot temperature including non-linear terms is obtained;

[0013] Perform local fitting on the non-linear terms, and combine the boundary conditions of the equivalent cylinder to obtain the linear mapping relationship between the local hot spot temperature, boundary conditions, and heat source distribution characteristics;

[0014] Introduce a compensation heat source correction term, introduce the compensation heat source into the traditional thermal network model in the form of an equivalent voltage source, and correct the prediction error caused by non-linear factors in the linearization process of the local hot spot temperature by adjusting the intensity of the compensation heat source, so as to construct a local hot spot temperature prediction model with a compensation heat source.

[0015] Furthermore, the improvement of the local hot spot prediction model of the motor is specifically as follows.

[0016] The boundary conditions include the radial inner wall temperature, the radial outer wall temperature, the axial front wall temperature, and the axial rear wall temperature.

[0017] Furthermore, the improvement of the local hot spot prediction model of the motor is specifically as follows.

[0018] Connect the local hot spot temperature prediction models with compensation heat sources of each cylindrical component according to the contact thermal resistance corresponding to the component heat transfer mode to obtain the local hot spot prediction model of the motor.

[0019] Furthermore, the local hot spot prediction model of the motor is as follows.

[0020] Based on the local hot spot prediction model of the motor, represent the heat exchange relationship between each node through the steady-state heat balance equation. The housing and the stator core are connected by a conduction thermal resistance, the shaft and the rotor are connected by a conduction thermal resistance, the end cover and the end cover air, the shaft and the end cover air, and the winding and the end cover air are connected by a convection thermal resistance. The stator core and the permanent magnet are connected by a conduction thermal resistance.

[0021] Furthermore, the solution of the steady-state heat balance equation, the calculation of the node temperature, and the obtaining of the local hot spot temperature of the corresponding component are specifically as follows.

[0022] According to the local hot spot prediction model of the motor, establish independent steady-state heat balance equations for each node to form a linear equation system, and solve the linear equation system to obtain the local hot spot temperature distribution of each component.

[0023] Furthermore, the steady-state heat balance equation is that, under steady-state conditions, the energy conservation of each thermal network node follows that the heat flowing in is equal to the heat flowing out. The heat flowing in includes the conduction heat and the heat source heat, and the heat flowing out includes the conduction and convection / radiation heat.

[0024] Furthermore, the specific form of the steady-state heat balance equation is as follows.

[0025]

[0026] Among them, Y is n*n the thermal conductivity matrix, and its elements represent the thermal conductivity between nodes; T is the node temperature matrix; PIt is a heat source matrix composed of core loss, winding copper loss and permanent magnet eddy current loss obtained by calculation in the finite element electromagnetic simulation model of the motor. C It is a compensation heat source matrix.

[0027] Furthermore, an iterative algorithm is used to adjust the compensation heat source matrix to obtain the local hot spot temperatures of each component of the motor, specifically including

[0028] Initialization: Assign the same initial temperature value to all unknown temperature nodes, and calculate the compensation heat source matrix according to the boundary conditions;

[0029] Substitute the compensation heat source matrix into the steady-state heat balance equation, calculate the node temperature distribution, and update the compensation heat source matrix according to the temperature distribution;

[0030] Alternately and cyclically execute the solution of the steady-state heat balance equation and the update of the compensation heat source matrix until the relative error percentage of the node temperature value is less than the preset convergence threshold, and it is determined that the convergence condition is reached.

[0031] The present invention discloses a high-precision prediction method for local hot spot temperature based on a compensation heat source, which relates to the technical field of motor simulation modeling. First, based on the steady-state heat conduction equation, a local hot spot temperature expression containing a non-linear term is derived. In the traditional thermal network model, the thermal conductivity of materials and the convective heat transfer coefficient are independent of temperature, and its heat balance equation shows a strictly linear form. However, the logarithmic term in the local hot spot temperature expression introduces non-linear characteristics, making the solution of the traditional thermal network model fail. Therefore, the least squares method is used to approximately express the temperature extreme value as a linear superposition of the boundary conditions and the heat source.

[0032] The present invention establishes a linear mathematical relationship model between the local hot spot temperature, the boundary conditions and the heat source based on the heat steady-state equation, and uses the lumped parameter thermal network method to calculate the local hot spot temperature of the permanent magnet motor, avoiding the high computational complexity of the traditional numerical method, and greatly improving the computational efficiency while ensuring the accuracy. In addition, the present invention innovatively introduces a compensation heat source in the form of a voltage source into the traditional thermal network model. Through this compensation mechanism, the fitting error generated during the non-linear to linear conversion of the local hot spot temperature is accurately corrected, thereby significantly improving the temperature prediction accuracy while maintaining the structural simplicity of the thermal network model. Generally speaking, the present invention not only significantly improves the prediction accuracy of the local hot spot temperature, but also greatly reduces the complexity of the heat conduction problem and improves the computational efficiency. Description of the Drawings

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on the provided drawings.

[0034] Figure 1 Flowchart of a high-precision prediction method for local hot spot temperature based on a compensation heat source disclosed by the present invention;

[0035] Figure 2 Motor local hot spot temperature model of the embodiment implemented by the present invention;

[0036] Figure 3 Equivalent hollow cylinder model related to the present invention;

[0037] Figure 4 Two equivalent thermal network models before and after improvement related to the present invention, where (a) is the traditional equivalent thermal network model and (b) is the equivalent thermal network model with a compensation heat source;

[0038] Figure 5 Flowchart of the motor temperature iterative calculation related to the present invention. Specific implementation manners

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0040] The present invention is based on two-dimensional thermal network modeling to improve the prediction accuracy of local hot spot temperature and reduce the complexity of heat conduction problems. Based on the principle of thermal steady-state balance, the present invention constructs a linear mathematical mapping model between local hot spot temperature, boundary conditions, and heat source distribution characteristics through theoretical derivation; further, a compensation heat source is introduced into the traditional thermal network model in the form of an equivalent voltage source, and by adjusting the intensity of the compensation heat source, the prediction error caused by non-linear factors in the linearization process of local hot spot temperature is effectively corrected. The present invention can be applied to fields such as low-speed high-torque permanent magnet synchronous motors, providing a scientific method that takes into account both accuracy and efficiency for the thermal management of high-torque density permanent magnet motors.

[0041] See Figure 1 As shown, the embodiments of the present invention disclose a high-precision prediction method for local hot spot temperature based on a compensation heat source, and the method includes:

[0042] S1: Build a finite element electromagnetic simulation model of the motor, and calculate the core loss, winding copper loss, and permanent magnet eddy current loss of the motor;

[0043] Specifically, in this embodiment, according to the topological structure and design parameters of the low-speed high-torque permanent magnet synchronous motor, as well as the material properties of the silicon steel sheet and the permanent magnet, a finite element electromagnetic simulation model of the motor is built. The finite element electromagnetic simulation model of the motor is an Ansys two-dimensional finite element electromagnetic simulation model, which is used to calculate the core loss, winding copper loss, and permanent magnet eddy current loss of the motor, and obtain the heat source distribution of the motor P .

[0044] S2: Divide and equivalent the motor into N cylindrical components according to the thermal conductivity of the material, where N is greater than or equal to 1; determine the contact thermal resistance between each component according to the heat transfer mode of the motor components;

[0045] Specifically, as shown in the appendix Figure 2 , in this embodiment, in order to consider the thermal network modeling of the motor, first, according to the thermal conductivity of the materials of each component of the motor, which is the thermal conductivity, it is divided into 5 cylindrical components with different geometric parameters, namely the stator core, permanent magnet, rotor core, and shaft in the radial direction. Among them, the stator core is further divided into stator teeth and stator yoke according to the heat source distribution.

[0046] Furthermore, in this embodiment, the division of each component of the motor also includes: equivalent simplification of heat dissipation and heat transfer in complex regions; considering the contact thermal resistance between different components: among them, heat conduction is considered between the housing and the stator core, and between the shaft and the rotor, and heat convection is considered between the end cover and the end cover air, and between the shaft and the end cover air. The connection of the thermal model of the motor components is mainly divided into conduction thermal resistance connection and convection thermal resistance connection according to the heat transfer mode.

[0047] S3: Based on each equivalent cylindrical component, construct a local hot spot temperature prediction model with a compensated heat source;

[0048] It should be noted that in this step, reasonable assumptions and simplifications are made according to the heat source distribution and heat transfer mechanism. Based on the equivalent cylindrical components in S2, a local hot spot temperature prediction model with a compensated heat source is constructed.

[0049] In this embodiment, in order to further improve the model accuracy, a compensated heat source term is introduced to quantify the error generated in the logarithmic term linearization process, and it is added to the traditional thermal network model in the form of an equivalent voltage source to construct a local hot spot temperature prediction model with high accuracy.

[0050] In this step, to derive the thermal resistance in the local hot spot temperature prediction model with a compensated heat source, reasonable assumptions and simplifications are made based on the heat source distribution and heat transfer mechanism. It is assumed that the radial and axial heat fluxes are independent of each other; a single average temperature is used to define the radial and axial heat fluxes, and the circumferential heat flux is ignored; the heat source is uniformly distributed in space; in this field, due to axial temperature symmetry, only half of the cylinder is modeled in the present invention.

[0051] In this embodiment, by performing a differential operation on the general solution of the radial steady-state heat conduction equation, the local hot spot temperature containing non-linear terms is obtained The expression is:

[0052] (1)

[0053] In Equation (1), c 1 and c 2 are constants; k represents the thermal conductivity, represents the heat source density in the equivalent hollow cylinder.

[0054] Perform a local fitting process on the non-linear term :

[0055] (2)

[0056] In Equation (2), f 1 and f 2 represent the coefficients in the linear fitting process.

[0057] Substitute Equation (2) into Equation (1), and combine with the boundary conditions of the equivalent cylinder, the radial inner wall temperature T r (r1), the radial outer wall temperature T r (r2) and the axial front wall temperature T a(0) , the axial rear wall temperature T a(L) , specifically see the appendix Figure 3 , the local hot spot temperature The linear mathematical mapping relationship between the boundary conditions and the heat source distribution characteristics can be obtained:

[0058] (3)

[0059] In Equation (3), r 1 and r 2 are the inner radius and outer radius of the cylinder respectively, L represents the axial length;

[0060] The non - linear term generates errors during the linear fitting process of Equation (2). To eliminate the influence of the errors on the model accuracy, a compensation heat source correction term is introduced. T comp :

[0061] (4)

[0062] In this embodiment, the present invention introduces the compensation heat source into the traditional thermal network model in the form of an equivalent voltage source. By adjusting the intensity of the compensation heat source, the prediction error caused by non - linear factors in the linearization process of the local hot - spot temperature is effectively corrected, and a local hot - spot temperature prediction model with a compensation heat source is established. The traditional thermal network model is improved, as shown in Figure 4 shown.

[0063]

[0064] S4: Improve the local hot - spot prediction model of the motor, solve the steady - state heat balance equation, calculate the temperature of each node, and obtain the local hot - spot temperature of the corresponding component.

[0065] In this embodiment, improving the local hot - spot prediction model of the motor includes connecting the established local hot - spot temperature models of each component with a compensation heat source according to the contact thermal resistance corresponding to the heat transfer mode of the component to form a local hot - spot prediction model of the permanent - magnet motor. Among them, between the motor housing and the stator core, between the shaft and the rotor are connected through conduction thermal resistance, between the end - cover and the end - cover air, between the shaft and the end - cover air, between the winding and the end - cover air are connected through convection thermal resistance, and between the stator core and the permanent magnet are connected through conduction thermal resistance. See Figure 2 shown, where the meanings of each thermal resistance are as follows.

[0066]

[0067] Furthermore, under steady - state conditions, the energy conservation of each thermal network node follows that the heat flowing in is equal to the heat flowing out. The thermal network is transformed into a thermal network corresponding to 16 nodes, and 16 independent steady - state heat balance equations are established to form a linear equation set. Solving the equation set can obtain the local hot - spot temperature distribution of each component. See the basic form of the (b) steady - state heat balance equation in Figure 4 which can be expressed as:

[0068]

[0069] Among them, Y is n*n the thermal conductivity matrix, and its elements represent the thermal conductivity between nodes; T is the node temperature matrix; P is the heat source matrix composed of the core loss, winding copper loss, and permanent - magnet eddy - current loss obtained from the electromagnetic finite - element model of the permanent - magnet motor in S1.C is a compensation heat source matrix.

[0070] Specifically, the inflowing heat includes conductive heat and heat source heat, and the outflowing heat includes conductive and convective / radiative heat.

[0071] In this embodiment, an iterative algorithm is used to adjust the compensation heat source matrix multiple times C to obtain the local hot spot temperatures of each component of the motor. The specific process is as shown in the appendix Figure 5 as follows. Specifically,

[0072] S41, Initialization: Assign the same initial temperature value to all unknown temperature nodes, and calculate the compensation heat source matrix according to the boundary conditions C ;

[0073] S42, Substitute the compensation heat source matrix into the steady-state heat balance equation to calculate the node temperature distribution, and update the compensation heat source matrix according to the temperature distribution;

[0074] S43, Alternately and cyclically execute the solution of the steady-state heat balance equation and the update of the compensation heat source matrix until the relative error percentage of the node temperature value is less than a preset convergence threshold, and determine that the convergence condition is reached.

[0075] The present invention establishes a linear mathematical relationship model between the local hot spot temperature, the boundary conditions, and the heat source based on the heat steady-state equation, and uses the lumped parameter thermal network method to calculate the local hot spot temperature of the permanent magnet motor, avoiding the high computational complexity of traditional numerical methods, and significantly improving the computational efficiency while ensuring accuracy. In addition, the present invention introduces a compensation heat source in the form of a voltage source into the traditional thermal network model, and precisely corrects the fitting error generated during the non-linear to linear conversion of the local hot spot temperature through this compensation mechanism, thereby significantly improving the temperature prediction accuracy while maintaining the structural simplicity of the thermal network model. The present invention not only significantly improves the prediction accuracy of the local hot spot temperature, but also greatly reduces the complexity of the heat conduction problem and improves the computational efficiency.

[0076] Corresponding to the above method embodiment, the embodiment of the present invention also provides a high-precision prediction system for local hot spot temperature based on a compensation heat source. A high-precision prediction system for local hot spot temperature based on a compensation heat source described below can be mutually referred to with a high-precision prediction method for local hot spot temperature based on a compensation heat source described above.

[0077] This system includes the following modules:

[0078] A high-precision prediction system for local hot spot temperature based on a compensation heat source, the system includes:

[0079] The system includes a loss calculation module, an equivalent module, a prediction model module, and a prediction solution module;

[0080] The loss calculation module is used to build a finite element electromagnetic simulation model of the motor and calculate the core loss, winding copper loss, and permanent magnet eddy current loss of the motor;

[0081] The equivalent module is used to divide and equivalent the motor into N cylindrical components according to the thermal conductivity of the material to obtain equivalent cylindrical components; N is greater than or equal to 1; determine the connecting thermal resistance between components according to the heat transfer mode of the motor components;

[0082] The prediction model module is used to build a local hot spot temperature prediction model with a compensated heat source;

[0083] The prediction solution module is used to improve the local hot spot prediction model of the motor, solve the steady-state heat balance equation, calculate the node temperature, and obtain the local hot spot temperature of the corresponding component.

[0084] Applying the system provided by the embodiment of the present invention, the loss calculation module builds a finite element electromagnetic simulation model of the motor and calculates the core loss, winding copper loss, and permanent magnet eddy current loss of the motor. The equivalent module divides and equivalent the motor into N cylindrical components according to the thermal conductivity of the material to obtain equivalent cylindrical components. The prediction model module builds a local hot spot temperature prediction model with a compensated heat source. The prediction solution module improves the local hot spot prediction model of the motor, solves the steady-state heat balance equation, calculates the node temperature, and obtains the local hot spot temperature of the corresponding component. Using the lumped parameter thermal network method to calculate the local hot spot temperature of the permanent magnet motor avoids the high computational complexity of the traditional numerical method and greatly improves the computational efficiency while ensuring the accuracy. In addition, the present invention innovatively introduces a compensated heat source in the form of a voltage source into the traditional thermal network model. Through this compensation mechanism, the fitting error generated in the conversion process from non-linearity to linearization of the local hot spot temperature is accurately corrected, thereby significantly improving the temperature prediction accuracy while maintaining the structural simplicity of the thermal network model. Generally speaking, the present invention not only significantly improves the prediction accuracy of the local hot spot temperature, but also greatly reduces the complexity of the heat conduction problem and improves the computational efficiency.

[0085] Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A high-precision prediction method for the temperature of local hot spots based on a compensated heat source, characterized in that An electromagnetic finite element simulation model of the motor is built to calculate the core loss, winding copper loss and permanent magnet eddy current loss of the motor, and the heat source distribution of the motor is obtained. The motor is divided into N equivalent cylindrical components according to the thermal conductivity of the material to obtain equivalent cylindrical components; N is greater than 1; according to the heat transfer mode of the motor components, the connecting thermal resistance between the components is determined. Based on each equivalent cylindrical component, a local hot spot temperature prediction model with a compensated heat source is constructed; specifically, by performing differential operations on the general solution of the radial steady-state heat conduction equation, a local hot spot temperature including non-linear terms is obtained; the non-linear terms are locally fitted, and combined with the boundary conditions of the equivalent cylinder, a linear mapping relationship between the local hot spot temperature, the boundary conditions and the heat source distribution characteristics is obtained; a compensated heat source correction term is introduced, and the compensated heat source is introduced into the traditional thermal network model in the form of an equivalent voltage source. By adjusting the intensity of the compensated heat source, the prediction error caused by non-linear factors in the linearization process of the local hot spot temperature is corrected, and a local hot spot temperature prediction model with a compensated heat source is constructed. The local hot spot temperature prediction models of each cylindrical component with a compensated heat source are connected according to the contact thermal resistance corresponding to the component heat transfer mode to obtain a local hot spot prediction model of the motor. According to the local hot spot prediction model of the motor, independent steady-state heat balance equations for each node are established to form a linear equation set, and the local hot spot temperature distribution of each component is obtained by solving the linear equation set. Among them, the specific form of the steady-state heat balance equation is Among them, Y is n*n a thermal conductivity matrix, whose elements represent the thermal conductivity between nodes; T the node temperature matrix; P is a heat source matrix composed of core loss, winding copper loss, and permanent magnet eddy current loss obtained from the finite element electromagnetic simulation model of the motor, C is the compensation heat source matrix; An iterative algorithm is used to adjust the compensated heat source matrix to obtain the local hot spot temperature of each component of the motor, including Initialization: The same initial temperature value is assigned to all unknown temperature nodes, and the compensated heat source matrix is calculated according to the boundary conditions. The compensated heat source matrix is substituted into the steady-state heat balance equation to calculate the node temperature distribution, and the compensated heat source matrix is updated according to the temperature distribution. The steady-state heat balance equation solution and the compensated heat source matrix update are alternately and cyclically executed until the relative error percentage of the node temperature value is less than the preset convergence threshold, and it is determined that the convergence condition is reached.

2. The high-precision prediction method for local hot spot temperature based on a compensation heat source according to claim 1, wherein The motor is divided and equivalent according to the thermal conductivity of the material into N cylindrical components. Specifically, the motor is divided into a stator core, a permanent magnet, a rotor core, and a rotating shaft according to the thermal conductivity of the material, and is equivalent to cylindrical components.

3. The high-precision prediction method for the local hot spot temperature based on the compensation heat source according to claim 1, characterized in that The boundary conditions include the radial inner wall temperature, the radial outer wall temperature, the axial front wall temperature and the axial rear wall temperature.

4. The high-precision prediction method for local hot spot temperature based on a compensation heat source according to claim 1, characterized in that, The local hot spot prediction model of the motor is Based on the local hot spot prediction model of the motor, the heat exchange relationship between each node is represented by the steady-state heat balance equation. The casing and the stator core are connected by a conduction thermal resistance, the shaft and the rotor are connected by a conduction thermal resistance, the end cover and the end cover air, the shaft and the end cover air, and the winding and the end cover air are connected by a convection thermal resistance. The stator core and the permanent magnet are connected by a conduction thermal resistance.

5. The high-precision prediction method for local hot spot temperature based on a compensation heat source according to claim 1, wherein The steady-state heat balance equation is that under steady-state conditions, the energy conservation of each thermal network node follows that the inflowing heat is equal to the outflowing heat. The inflowing heat includes the conduction heat and the heat source heat, and the outflowing heat includes the conduction and convection / radiation heat.

Images